“Let’s work on getting those blood sugars down, honey.”
On the heels of a report finding no association between Alzheimer’s disease and abnormal blood sugar metabolism, MedPageToday features an new study linking high blood sugars to future development of dementia. And diabetics with sugar levels higher than other diabetics were more prone to develop dementia.
Some of you have already noted that not all cases of dementia are Alzheimer’s dementia. But Alzheimer’s accounts for a solid majority of dementia cases.
Some quotes from MedPageToday:
During a median follow-up of 6.8 years, 524 participants [of the 2000 total] developed dementia, consisting of 74 with diabetes and 450 without. Patients without diabetes and who developed dementia had significantly higher average glucose levels in the 5 years before diagnosis of dementia (P=0.01). The difference translated into a hazard ratio of 1.18 (95% CI 1.04-1.33).
…
Among the patients with diabetes, glucose levels averaged 190 mg/dL in those who developed dementia versus 160 mg/dL in those who did not. The difference represented a 40% increase in the hazard for dementia (HR 1.40, 95% CI 1.12-1.76).
Reference: Crane PK et al. “Glucose Levels and Risk of Dementia” N Engl J Med 2013; 369: 540-548.
Reminder: Conquer Diabetes and Prediabetes is now available on Kindle.
Advanced Mediterranean Diet 
I haven’t been to your site in a while! Glad you are still fighting the good fight for a healthier world, Steve!
Good to hear from you, Dr. J! I gotta visit your blog and CalorieLab again. Just got out of the habit.
-Steve
Steve – I read in another post that you mentioned thinking the link between blood glucose and dementia may be weak with all of these conflicting results. Looking at the studies, do you think any one or two are more representative of the relationship than the others? I imagine these studies vary in quality and assessment tools.
Lately, I’m thinking there is indeed a definite link between diabetes and dementia. I’ll never know how much publication bias is involved; i.e., studies that show no association aren’t as exciting, so may not be published.